This preview shows page 1. Sign up to view the full content.
Unformatted text preview: . The estimated marginal effect is
The second question calls for an F-test.
Since the critical value at the 5 percent level is 3.00, we strongly reject.
3. We have (
( ̂) ( ̂) (( ̂ )
() ( ̂ )) ) () )
() (( (( ) ( ( )
() (( ) ( ) ( )) 4. In the first regression the Durbin-Watson isn’t very high. While the table doesn’t go up
to as many observations as we have, you would probably reject no serial
correlation…but not very strongly. In contrast, the Breusch -Godfrey test strongly rejects
no serial correlation. Since there is a lagged dependent variable, we need to use the
View Full Document